A variety of techniques have been traditionally employed to facilitate the preparation of intracellular proteins from microorganisms. Typically, the initial steps in these techniques involve lysis, rupture or permeabilization of the bacterial cells, to disrupt the bacterial cell wall and allow release of the intracellular proteins into the extracellular milieu. Following this release, the desired proteins are purified from the extracts, typically by a series of chromatographic steps.
Several approaches have proven useful in accomplishing the release of intracellular proteins from bacterial cells. Included among these are the use of chemical lysis or permeabilization, physical methods of disruption, or a combination of chemical and physical approaches (reviewed in Felix, H., Anal. Biochem. 120:211-234 (1982)).
Chemical methods of disruption of the bacterial cell wall that have proven useful include treatment of cells with organic solvents such as toluene (Putnam, S. L., and Koch, A. L., Anal. Biochem. 63:350-360 (1975); Laurent, S. J., and Vannier, F. S., Biochimie 59:747-750 (1977); Felix, H., Anal. Biochem. 120:211-234 (1982)), with chaeotropes such as guanidine salts (Hettwer, D., and Wang, H., Biotechnol. Bioeng. 33:886-895 (1989)), with antibiotics such as polymyxin B (Schupp, J. M., et al., BioTechniques 19:18-20 (1995); Felix, H., Anal. Biochem. 120:211-234 (1982)), or with enzymes such as lysozyme or lysostaphin (McHenry, C. S., and Kornberg, A., J. Biol. Chem. 252(18):6478-6484 (1977); Cull, M., and McHenry, C. S., Meth. Enymol. 182:147-153 (1990); Hughes, A. J., Jr., et al., J Cell. Biochem. Suppl. 0 16(Part B):84 (1992); Sambrook, J., et al., in: Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, p. 17.38 (1989); Ausubel, F. M., et al., in: Current Protocols in Molecular Biology, New York: John Wiley & Sons, pp. 4.4.1-4.4.7 (1993)). The permeabilization effects of these various chemical agents may be enhanced by concurrently treating the bacterial cells with detergents such as TRITON X-100.RTM., sodium dodecylsulfate (SDS) or Brij 35 (Laurent, S. J., and Vannier, F. S., Biochimie 59:747-750 (1977); Felix, H., Anal. Biochem. 120:211-234 (1982); Hettwer, D., and Wang, H., Biotechnol. Bioeng. 33:886-895 (1989); Cull, M., and McHenry, C. S., Meth. Enzymol. 182:147-153 (1990); Schupp, J. M., et al., BioTechniques 19:18-20 (1995)), or with proteins or protamines such as bovine serum albumin or spermidine (McHenry, C. H., and Kornberg, A., J. Biol. Chem. 252(18):6478-6484 (1977); Felix, H., Anal. Biochem. 120:211-234 (1982); Hughes, A. J., Jr., et al., J. Cell. Biochem. Suppl. 0 16(Part B):84 (1992)).
In addition to these various chemical treatments, a number of physical methods of disruption have been used. These physical methods include osmotic shock, e.g., suspension of the cells in a hypotonic solution in the presence or absence of emulsifiers (Roberts, J. D., and Lieberman, M. W., Biochemistry 18:4499-4505 (1979); Felix, H., Anal. Biochem. 120:211-234 (1982)), drying (Mowshowitz, D. B., Anal. Biochem. 70:94-99 (1976)), bead agitation such as ball milling (Felix, H., Anal. Biochem. 120:211-234 (1982); Cull, M., and McHenry, C. S., Meth. Enzymol. 182:147-153 (1990)), temperature shock, e.g., freeze-thaw cycling (Lazzarini, R. A., and Johnson, L. D., Nature New Biol. 243:17-20 (1975); Felix, H., Anal. Biochem. 120:211-234 (1982)), sonication (Amos, H., et al., J. Bacteriol. 94:232-240 (1967); Ausubel, F. M., et al., in: Current Protocols in Molecular Biology, New York: John Wiley & Sons, pp. 4.4.1-4.4.7 (1993)) and pressure disruption, e.g., use of a french pressure cell (Ausubel, F. M., et al., in: Current Protocols in Molecular Biology, New York: John Wiley & Sons, pp. 16.8.6-16.8.8 (1993)). Other approaches combine these chemical and physical methods of disruption, such as lysozyme treatment followed by sonication or pressure treatment, to maximize cell disruption and protein release (Ausubel, F. M., et al., in: Current Protocols in Molecular Biology, New York: John Wiley & Sons, pp. 4.4.1-4.4.7 (1993)).
These disruption approaches have several advantages, including their ability to rapidly and completely (in the case of physical methods) disrupt the bacterial cell such that the release of intracellular proteins is maximized. In fact, these approaches have been used in the initial steps of processes for the purification of a variety of bacterial cytosolic enzymes, including natural and recombinant proteins from mesophilic organisms such as Escherichia coli, Bacillus subtilis and Staphylococcus aureus (Laurent, S. J., and Vannier, F. S., Biochimie 59:747-750 (1977); Cull, M., and McHenry, C. S., Meth. Enzymol. 182:147-153 (1990); Hughes, A. J., Jr., et al., J. Cell. Biochem. Suppl. 0 16(Part B):84 (1992); Ausubel, F. M., et al., in: Current Protocols in Molecular Biology, New York: John Wiley & Sons, pp. 4.4.1-4.4.7 (1993)), as well as phosphatases, restriction enzymes, DNA or RNA polymerases and other proteins from thermophilic bacteria and archaea such as Thermus aquaticus, Thermus thermophilus, Thermus flavis, Thermus caldophilus, Thermotoga maritima, and Sulfolobus acidocaldarius (Shinomiya, T., et al., J. Biochem. 92(6):1823-1832 (1982); Elie, C., et al., Biochim. Biophys. Acta 951(2-3):261-267 (1988); Palm, P., et al, Nucl. Acids Res. 21(21):4904-4908 (1993); Park, J. H., et al., Eur. J. Biochem. 214(1):135-140 (1993); Harrell, R. A., and Hand, R. P., PCR Meth. Appl. 3(6):372-375 (1994); Meyer, W., et al., Arch. Biochem. Biophys. 319(1):149-156 (1995)).
However, these methods possess distinct disadvantages as well. For example, the physical methods by definition involve shearing and fracturing of the bacterial cell walls and plasma membranes. These processes thus result in extracts containing large amounts of particulate matter, such as membrane or cell wall fragments, which must be removed from the extracts, typically by centrifugation, prior to purification of the enzymes. This need for centrifugation limits the batch size capable of being processed in a single preparation to that of available centrifuge space; thus, large production-scale preparations are impracticable if not impossible. Furthermore, physical methods, and many chemical permeation techniques, typically result in the release from the cells not only of the desired intracellular proteins, but also of undesired nucleic acids and membrane lipids (the latter particularly resulting when organic solvents are used to permeabilize the cells). These undesirable cellular components also complicate the subsequent processes for purification of the desired proteins, as they increase the viscosity of the extracts (Sambrook, J., et al., in: Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, p. 17.38 (1989); Cull, M., and McHenry, C. S., Meth. Enzymol. 182:147-153 (1990)), and bind with high avidity and affinity to nucleic acid-binding proteins such as DNA polymerases, RNA polymerases and restriction enzymes.
These limitations have been partially overcome in the case of proteins prepared from mesophilic bacteria. For example, mild chemical disruption of E. coli, B. subtilis and Salmonella typhimurium has been conducted to permeabilize these cells, allowing free mobility of proteins across the membrane of the cells or resultant spheroplasts, but inducing retention of most of the nucleic acids within the cell or spheroplast (Laurent, S. J., and Vannier, F. S., Biochimie 59:747-750 (1977); Hettwer, D., and Wang, H., Biotechnol. Bioeng. 33:886-895 (1989); Cull, M., and McHenry, C. S., Meth. Enzymol. 182:147-153 (1990); Schupp, J. M., et al., BioTechniques 19:18-20 (1995)). Similar approaches have also been taken to limit the contamination of protein preparations from thermophilic bacteria, based on the demonstration that protein-permeable spheroplast can be prepared from thermophiles like Thermus thermophilus by treatment with lysozyme with or without a detergent (Oshima, T., and Imahori, K., Int. J. Syst. Bacteriol. 24(1):102-112 (1974)).
These approaches, however, are insufficient for preventing the contamination of preparations of thermostable enzymes with DNA. For example, it has been reported by at least two different groups that commercially available preparations of Taq DNA polymerase are contaminated with bacterial DNA (Rand, K. H., and Houck, H., Mol. Cell Probes 4(6):445-450 (1990); Hughes, M. S., et al., J. Clin. Microbiol. 32(8):2007-2008 (1994)), despite the use of gentle lysis procedures to liberate the enzyme from the cells. Furthermore, this contaminating DNA may come not only from the organisms which are the source of the enzyme (Thermus aquaticus for natural Taq polymerase; E. coli for recombinant enzyme), but also from unknown organisms present in the reagents and materials used to purify the enzyme after its release from the cells (Rand, K. H., and Houck, H., Mol. Cell Probes 4(6):445-450 (1990); Hughes, M. S., et al., J. Clin. Microbiol. 32(8):2007-2008 (1994)). Since thermophilic enzymes such as DNA polymerases and restriction enzymes are routinely used in automated techniques of DNA amplification and sequencing, e.g., the Polymerase Chain Reaction (PCR), the presence of contaminating DNA in the enzyme preparations is a significant problem since it can give rise to spurious amplification or sequencing results. Thus, a need exists for preparations of thermostable enzymes that are substantially free of contamination by nucleic acids.
Various solutions to this problem have been suggested. For example, many investigators routinely run "no-template controls" to allow subtraction of any spurious results from their experimental samples, or to determine the extent of contamination of their enzyme preparations ((Rand, K. H., and Houck, H., Mol. Cell Probes 4(6):445-450 (1990)). This approach, however, increases the resource and time requirements, and thus the expense, of the assays, particularly in high-throughput settings such as in clinical applications of PCR. Also suggested have been methods of eliminating nucleic acids in the enzyme preparations, such as treatment of the preparations with DNAse or RNAse, restriction enzyme digestion, organic phase partitioning, or cesium chloride density gradient centrifugation (Cull, M., and McHenry, C. S., Meth. Enzymol 182:147-153 (1990); Rand, K. H., and Houck, H., Mol. Cell Probes 4(6):445-450 (1990)), although these approaches have apparently not proved routinely successful at removing contaminating DNA (Rand, K. H., and Houck, H., Mol. Cell Probes 4(6):445-450 (1990)). Other methods of inactivating DNA have been described, such as a method of heating samples at 60.degree.-100.degree. C. in the presence of an acid at pH 3-4 (U.S. Pat. No. 5,417,862). However, while thermophilic enzymes are fairly resistant to these increased temperatures, they quickly lose enzymatic activity when exposed to a pH below about 5, thus precluding use of this method in purging thermophilic enzyme preparations of nucleic acid contamination.
Thus, instead of attempting to remove nucleic acids from preparations of thermostable enzymes, a more reasonable and successful approach would be to prevent contamination of the enzymes by nucleic acids from the outset in the purification process. Such an approach would be two-pronged: 1) preventing release of nucleic acids from the bacterial cells during permeabilization of the cells to release the enzymes; and 2) preventing contamination of the enzymes during the purification process itself. Furthermore, an optimal method would obviate the need for centrifugation in the process, thus allowing large-scale, and even continuous, production of nucleic acid-free thermophilic enzymes. The present invention provides such methods, and thermophilic enzymes produced by these methods.